update benchmark

alphamind/benchmarks/benchmarks.py

@@ -18,21 +18,22 @@ if __name__ == '__main__':
 
     benchmark_neutralize(3000, 10, 1000)
     benchmark_neutralize(30, 10, 50000)
-    benchmark_standardize(3000, 10, 1000)
-    benchmark_standardize_with_group(3000, 10, 1000, 30)
-    benchmark_standardize(30, 10, 50000)
-    benchmark_standardize_with_group(30, 10, 5000, 5)
-    benchmark_standardize(50000, 50, 20)
-    benchmark_standardize_with_group(50000, 50, 20, 50)
-    benchmark_winsorize_normal(3000, 10, 1000)
-    benchmark_winsorize_normal_with_group(3000, 10, 1000, 30)
-    benchmark_winsorize_normal(30, 10, 50000)
-    benchmark_winsorize_normal_with_group(30, 10, 5000, 5)
-    benchmark_winsorize_normal(50000, 50, 20)
-    benchmark_winsorize_normal_with_group(50000, 50, 20, 50)
-    benchmark_build_rank(3000, 1000, 300)
-    benchmark_build_rank_with_group(3000, 1000, 10, 30)
-    benchmark_build_rank(30, 50000, 3)
-    benchmark_build_rank_with_group(30, 50000, 1, 3)
-    benchmark_build_rank(50000, 20, 3000)
-    benchmark_build_rank_with_group(50000, 20, 10, 300)
+    benchmark_neutralize(50000, 50, 20)
+    # benchmark_standardize(3000, 10, 1000)
+    # benchmark_standardize_with_group(3000, 10, 1000, 30)
+    # benchmark_standardize(30, 10, 50000)
+    # benchmark_standardize_with_group(30, 10, 5000, 5)
+    # benchmark_standardize(50000, 50, 20)
+    # benchmark_standardize_with_group(50000, 50, 20, 50)
+    # benchmark_winsorize_normal(3000, 10, 1000)
+    # benchmark_winsorize_normal_with_group(3000, 10, 1000, 30)
+    # benchmark_winsorize_normal(30, 10, 50000)
+    # benchmark_winsorize_normal_with_group(30, 10, 5000, 5)
+    # benchmark_winsorize_normal(50000, 50, 20)
+    # benchmark_winsorize_normal_with_group(50000, 50, 20, 50)
+    # benchmark_build_rank(3000, 1000, 300)
+    # benchmark_build_rank_with_group(3000, 1000, 10, 30)
+    # benchmark_build_rank(30, 50000, 3)
+    # benchmark_build_rank_with_group(30, 50000, 1, 3)
+    # benchmark_build_rank(50000, 20, 3000)
+    # benchmark_build_rank_with_group(50000, 20, 10, 300)

alphamind/benchmarks/data/neutralize.py

@@ -10,7 +10,7 @@ import datetime as dt
 import numpy as np
 from sklearn.linear_model import LinearRegression
 
-from alphamind.data.neutralize import ls_fit
+from alphamind.data.neutralize import neutralize
 
 
 def benchmark_neutralize(n_samples: int, n_features: int, n_loops: int) -> None:
@@ -23,7 +23,7 @@ def benchmark_neutralize(n_samples: int, n_features: int, n_loops: int) -> None:
 
     start = dt.datetime.now()
     for _ in range(n_loops):
-        _ = ls_fit(x, y)
+        _ = neutralize(x, y)
     impl_model_time = dt.datetime.now() - start
 
     print('{0:20s}: {1}'.format('Implemented model', impl_model_time))
@@ -31,7 +31,8 @@ def benchmark_neutralize(n_samples: int, n_features: int, n_loops: int) -> None:
     start = dt.datetime.now()
     for _ in range(n_loops):
         benchmark_model = LinearRegression(fit_intercept=False)
-        _ = benchmark_model.fit(x, y)
+        benchmark_model.fit(x, y)
+        _ = y - x @ benchmark_model.coef_
     benchmark_model_time = dt.datetime.now() - start
 
     print('{0:20s}: {1}'.format('Benchmark model', benchmark_model_time))

alphamind/data/neutralize.py

@@ -15,7 +15,7 @@ def neutralize(x: np.ndarray, y: np.ndarray) -> np.ndarray:
 
 
 def ls_fit(x: np.ndarray, y: np.ndarray) -> np.ndarray:
-    x_bar = np.transpose(x)
+    x_bar = x.T
     b = solve(x_bar @ x, x_bar @ y)
     return b
 

alphamind/portfolio/rankbuilder.py

@@ -6,15 +6,17 @@ Created on 2017-4-26
 """
 
 import numpy as np
+from numpy import zeros
+from numpy import max
 
 
 def rank_build(er: np.ndarray, use_rank: int, groups: np.ndarray=None) -> np.ndarray:
     neg_er = -er
-    masks = np.zeros(len(er), dtype=bool)
+    masks = zeros(len(er), dtype=bool)
     ordering = neg_er.argsort()
 
     if groups is not None:
-        max_g = np.max(groups)
+        max_g = max(groups)
         index_range = np.arange(len(er))
 
         for i in range(max_g + 1):
@@ -23,11 +25,10 @@ def rank_build(er: np.ndarray, use_rank: int, groups: np.ndarray=None) -> np.nda
             current_ordering = neg_er[current_mask].argsort()
             masks[current_index[current_ordering[:use_rank]]] = True
     else:
-
         masks[ordering[:use_rank]] = True
 
-    weights = np.zeros(len(er))
-    weights[masks] = 1. / np.sum(masks)
+    weights = zeros(len(er))
+    weights[masks] = 1. / use_rank
     return weights
 
 
@@ -43,3 +44,4 @@ if __name__ == '__main__':
     for i in range(10000):
         weights = rank_build(x, 30, groups)
     print(dt.datetime.now() - start)
+